<<

Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

Thorax (1973), 28, 379.

A clinical and radiological review of 204 hiatal operations E. HOFFMAN and M. C. SUMNER

Department of Thoracic , Poole Hospital, Nunthorpe, Middlesbrough

Two hundred and four patients with reflux oesophagitis and hiatal hernia were operated on by a modified Allison's technique and followed up for one to 10 years. To avoid surgical bias, the clinical and radiological findings were assessed by the co-author, a radiologist. Ages, type of hernia, symptoms, coexisting pathology, and operative technique are described. The late follow-up includes an assessment of patients' opinions of their operation and residual symptoms and a review of the radiological findings. The management of 25 patients with fibrous strictures is described. Factors preventing reflux are discussed. Current operative procedures for reflux oesophagitis are reviewed. In this series of 204 cases, five patients, that is 2 5%, were surgical failures. They were all dissatisfied with their operation and their symptoms had not improved; radiologically three of them showed reflux or an irreducible hiatal hernia. This failure rate is not considered unreasonable and the modified Allison's procedure is therefore recommended for patients with reflux oesophagitis.

There is still a wide divergence of opinion about TABLE II the management of hiatal hernia and gastro- AGE DISTRIBUTION http://thorax.bmj.com/ oesophageal reflux. A variety of operations have been devised and discarded and there is still no Age at Operation Men Women Total % 20-30 1 1 2 0 9 standard procedure. Allison's (1951) operation 30-40 11 5 16 7-8 was the first to be based on sound physiological 40-50 28 17 45 22-1 50-60 22 49 71 34-8 principles, and one of us (E.H.) has been using 60-70 10 55 65 31-9 a modified Allison's procedure for nearly 20 years. 70-80 2 3 5 2 5 Total 74 130 204 100 CLINICAL MATERIAL on September 23, 2021 by guest. Protected copyright. In order to evaluate this treatment we decided to review all hiatal hernia cases which were operated on TABLE III by one of us (E.H.) between 1959 and 1969 with a TYPE OF HERNIA follow-up period of one to 10 years (Table I). In all, there were 232 patients with three postoperative deaths No. % (two from pulmonary emboli and one from uraemia). Sliding hiatus hernia 185 90 8 Paraoesophageal hernia 6 2-9 Fourteen died at home of causes not connected with Recurrent hernia 6 2-9 their operation and 11 could not be traced or refused Reflux oesophagitis (no hernia demonstrated) 7 3.4 to attend for follow-up. Most surgeons review their own results, and to avoid bias, in this series all patients were assessed both AGES The series included 74 men and 130 women. clinically and radiologically by the co-author (M.C.S.), Most women were in the 50-70 age group, while men a radiologist. were mainly between 40 and 60 (Table II). Over two- thirds of the men did heavy work. TABLE I TYPE OF HERNIA In this series, sliding hiatal herniae OPERATIONS FOR HIATUS HERNIA 1959-69 were by far the commonest (185) (Table III). There Total cases 232 were six paraoesophageal herniae, all of which at Died in hospital 3 operation were found to be of a mixed type with the Died at home 14 Untraceable or refused to attend 11 cardia above the diaphragm. Of the six recurrent Patients reviewed 204 herniae, four had previously been repaired through the by general surgeons and the other two by the 379 Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

380 E. Hoffman and M. C. Sumner author (E.H.) through a thoracic incision. Severe TABLE VI symptoms of reflux oesophagitis requiring surgery may PREVIOUS OPERATIONS also be present without a demonstrable hernia (Hiebert and Belsey, 1961); this occurred in seven patients. Appendicectomy 17 Hysterectomy 15 Varicose veins 15 SYMPTOMS Patients with sliding and paraoesophageal Uterine prolapse 9 herniae complained of symptoms of reflux oesophagitis Cholecystectomy 8 Partial gastrectomy or gastroenterostomy 6 (Table IV). The most frequent symptoms were retro- 6 sternal or epigastric pain (117) and acid regurgitation Thyroidectomy 4 Oophorectomy 4 (102). Symptoms were often aggravated by change of Haemorrhoidectomy 4 posture (93). Sickness or were present in 70 patients. Sixty-five complained of and of these, 25 had a fibrous stricture. Fifty-six patients com- herniae. Casten, Bernhang, Nach, and Spinzia (1963) plained of , 49 of , and 30 lost stated that duodenal ulceration was present in 50% of weight. Haematemesis occurred in 18 patients and was their patients with symptomatic hiatal herniae. Brain usually mild. Other symptoms were-dyspnoea in (1966), however, found an incidence of only 4-2%, and eight, pain between the shoulders in eight, and Collis (1970) had an incidence below 6-5%. In our melaena in two cases. series the preoperative barium meal showed only one patient with a duodenal ulcer although six others stated that they had had peptic ulceration in the past. TABLE IV Twelve patients had a blood pressure of PREOPERATIVE SYMPTOMS 200/110 mmHg or over. Eighteen patients had chronic bronchitis and four suffered from ischaemic heart No. Patients % disease. Severe kyphoscoliosis was present in four Retrosternal or epigastric pain 117 57 4 cases. Acid regurgitation 102 50-0 Postural aggravation 93 456 Some patients had had previous surgical treatment Sickness or vomiting 70 34-3 (Table VI). Dysphagia 65 31 8 Heartburn 56 27 5 Indigestion (flatulence, belching or OPERATIVE TECHNIQUE epigastric discomfort) 49 24-0 Loss of weight 30 14 7 Haematemesis 18 8.8 Preoperative oesophagoscopy was carried out in http://thorax.bmj.com/ Dyspnoea 8 3 9 Pain between shoulders 8 3-9 all cases to assess the degree of oesophagitis and Melaena 2 0-9 to exclude other pathology. A modified Allison's repair was performed on all patients. The chest was opened through the TABLE V seventh interspace. Early in the series two to three intercostal nerves were divided to prevent post. ANAEMIA ON ADMISSION operative pain, but this was abandoned because Hb % No. Patients % of troublesome numbness and bulging of the left on September 23, 2021 by guest. Protected copyright. 30-40 5 2-50 epigastrium. 40-50 1 0-49 50-60 4 1 90 The oesophagus was mobilized up to the in- 60-70 7 3 40 ferior pulmonary vein or higher to prevent ten- sion after reduction of the cardia below the diaphragm. Care was taken not to divide any Preoperative anaemia was present in 17 patients vagal fibres during dissection of the oesophagus (Table V). Windsor and Collis (1967) stated that because vagal damage was thought to be the cause anaemia occurred three times as often in paraoeso- of transient diarrhoea in three of our patients. phageal herniae as in the sliding variety. In our A finger of the left hand was passed through patients with anaemia, one had a paraoesophageal hernia and 16 had herniae of the sliding variety. a short diaphragmatic incision and through the Two of the paraoesophageal herniae were admitted hiatus into the hernial sac. The sac was not divided, with acute obstructive symptoms and were operated on as in Allison's repair, but was left intact and a after the acute stage had subsided. These herniae purse string suture was put around its periphery should be repaired, even if symptom free, because of close to but not including the oesophagus. The the danger of strangulation (Hoffman, 1968). purse string suture and a tape around the oeso- In two of the six recurrent herniae there was no phagus were then reduced below the diaphragm. postoperative relief of symptoms and the operation The purse string was then tied and stitched to must be presumed to have been ineffective. the undersurface of the diaphragm. Two or three oTHmR PATHOLOGY Opinions vary about how often interrupted stitches were usually added to re- peptic ulceration occurs in patients with hiatal inforce the fixation of the sac to the diaphragm. Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

A clinical and radiological review of 204 hiatal hernia operations 381

After the limbs of the right crus were dissected, nature. In two patients it was severe enough to the diaphragmatic reflection of the pericardium require intercostal neurectomy. was mobilized. The first retro-oesophageal suture approximating the limbs of the crus included a LATE FOLLOW-UP good bite of the tendinous diaphragm medially. In closing the diaphragmatic incision care was All patients attended personally for follow-up. taken to include both the pleural and peritoneal They were asked how satisfied they were with layers. The object of this was to prevent herniation their operation and were questioned about any and strangulation of the through the symptoms. A barium meal was done and was diaphragmatic incision, as described by Effier compared with the pre- and postoperative barium (1965). films. We did not do special studies such as intra- Allison's technique was followed except for two luminal pressure and pH recordings. modifications. First, the sac was preserved and PATIENTS' ASSESSMENT Of the 204 patients, 155 fixed to the diaphragm with a purse string, be- (75 9%) were completely satisfied, 37 (18-1 %) cause we believe that this is a more secure method were partially satisfied, and 12 (59%) were dis- of fixation of the cardia below the diaphragm. satisfied with their operation (Table VII). The Secondly, in our opinion the inclusion of the proportion of those completely satisfied and par- tendinous portion of the diaphragm in the first tially satisfied was similar in both sexes; there retro-oesophageal suture plays an important part were, however, relatively more dissatisfied women in preventing recurrence. than men. POSTOPERATIVE COMPLICATIONS TABLE VII The most frequent postoperative was PATIENT'S ASSESSMENT OF RESULT pulmonary embolism which occurred in 12 of our Men Women Total % patients. In recent years all patients except the Completely satisfied 58 97 155 759 elderly have been given anticoagulants from the Partially satisfied 14 23 37 18.1 fourth postoperative day, although their value in Dissatisfied 2 10 12 5-9 http://thorax.bmj.com/ preventing and treating postoperative thrombosis and embolism remains controversial. RESIDUAL SYMPTOMS One hundred and thirty-six Persistent gastric dilatation was present in three (66'7%) patients had no symptoms of oesophag- patients. One of these had a past history of pyloric itis (Table VIII). In 61 (29-9%) symptoms were stenosis; in the other two no obvious pathology either mild or moderate. Seven (3-4%) patients could be found. One of these had such severe said that their symptoms had not improved. dilatation that laparotomy became necessary. Symptoms of oesophagitis were thus either cured Nothing abnormal was found, but the dilatation

or improved in 197 (96-6%). on September 23, 2021 by guest. Protected copyright. persisted in spite of a gastroenterostomy. The In 15 of the 49 partially satisfied and dissatisfied patient's condition improved only after large patients symptoms were due to coincidental patho- doses of potassium, although the serum electro- logical conditions, such as coronary thrombosis, lytes were within normal limits. hypertension, asthmatic bronchitis, osteoarthritis Two patients had erythema due to anticoagu- of the spine, and psychosis. lants, and two developed auricular fibrillation. In another case the thoracotomy incision broke RADIOLOGICAL FINDINGS The technique employed down and the wound had to be resutured. by radiologists to demonstrate reflux or a hernia Incisional pain is a well-recognized complica- varies. In this series patients were examined in tion of low thoracic incisions. It occurred in 26 the upright position while drinking barium, and of our patients but was usually of a transient they were then tilted into the Trendelenburg posi-

TABLE VIII RESIDUAL SYMPTOMS OF OFSOPHAGITIS AND PATIENT'S ASSESSMENT OF RESULT Completely Satisfied Partially Satisfied Dissatisfied Symptoms of oesophagitis None Mild None Mild or improved No change None Improved No change 121 34 10 24 3 S5 3 4 Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

382 E. Hoffman and M. C. Sumner TABLE IX RESIDUAL REFLUX AND HERNIA AND ITS RELATION TO PATIENT'S ASSESSMENT AND SYMPTOMS OF OESOPHAGITIS Completely Satisfied | Partially Satisfied Dissatisfied Total Symptoms None Mild Improved No Change Improved No Change Reflux only 7 4 1 0 0 1 13 Reflux+hiatus hernia (in Trendelenburg position) 1 4 1 1 0 0 7 Reflux+irreducible hiatus hernia 0 0 0 0 1 1 2 Total 8 8 2 1 1 2 22(108%)

TABLE X LATE ABNORMAL BARIUM MEAL APPEARANCES AND RELATION TO PATIENT'S ASSESSMENT OF RESULT Completely Satisfied Partially Satisfied Dissatisfied Total % Reflux or reflux and hernia 17 3 2 22 10-8 Delay of passage of barium due to: (a) spasm and incoordination ofmovements 5 4 2 11 54 (b) residual stricture 3 1 0 4 1-9 Diverticulum of oesophagus or stomach 7 1 1 9 4.4 Partial torsion of stomach 4 3 0 7 3*4 Gastric or duodenal ulcer 1 1 0 2 09 Partial gastrectomy (Billroth I) 2 0 1 3 1-4

tion lying supine. They were asked to turn un- All 22 patients who had reflux or herniation aided on to their right side and told to developed it either immediately postoperatively or several times in quick succession and to swallow. during the first two years. http://thorax.bmj.com/ If the slightest reflux was seen they were then In 15 patients there was delayed passage of asked to raise their legs. After films were taken barium at the lower end of the oesophagus (Table in the right Trendelenburg position patients were X). In four this was due to a residual stricture. told to turn unaided on to their back and then to In 11 cases there was spasm and incoordination the left side. Films were then taken in the left of the oesophageal wall. This has been ascribed oblique Trendelenburg position. to reflux oesophagitis by Olsen and Schlegel Postoperative reflux was seen in 22 patients (1965), but in our cases reflux could be demon- (Table IX). In 13 of these reflux alone was pre- strated in only two patients. Of the remaining sent, in the remaining nine a hiatal hernia was nine, abnormal peristalsis in three patients was on September 23, 2021 by guest. Protected copyright. also found. Seven of the hiatal herniae appeared thought to be due to low oesophageal diverticula only in the Trendelenburg position and two were and in the other six no pathology was found irreducible. The demonstration of reflux alone or although the hiatal repair might have been too reflux with a reducible hiatal hernia did not affect tight. patients' opinions of their operation; 16 of the Of the nine diverticula, seven were oesophageal 22 patients with reflux were completely satisfied. and two gastric. The oesophageal diverticula were Mild or moderate symptoms, however, were pre- small and found just above the diaphragm. Seven sent in 11 of the 22 cases. Two patients with reflux patients had partial torsion of the stomach; three and an irreducible hiatal hernia were both dis- of these had originally had large paraoesophageal satisfied and their symptoms persisted. Although herniae. it is not known why some people with reflux Peptic ulceration was rare at follow-up; only develop oesophagitis and others do not, it is one patient had a gastric ulcer and one a duo- likely that the amount of reflux matters. The type denal, although postoperatively peptic ulceration of reflux is also important. Reflux of bile leads to was seen in seven patients. a more severe oesophagitis than that with acid Billroth I gastrectomy is not uncommonly asso- pepsin regurgitation (Moffat and Berkas, 1965). ciated with reflux oesophagitis (Windsor, 1964). This was observed in three patients who developed Three of our patients had already had a Billroth a severe oesophagitis following a Billroth I I gastrectomy. In two of these, reflux symptoms gastrectomy. appeared in the early postoperative period. One Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

A clinical and radiological review of 204 hiatal hernia operations 383 had a prolonged dilatation of the gastric remnant DISCUSSION postoperatively, which may have facilitated reflux of bile previously described by Toye and Williams It has been accepted in recent years that there are two main factors in the control of reflux: (1965). (a) a normal anatomical position of the intra- abdominal segment of the oesophagus, and (b) an STRICTURES effective internal oesophageal sphincter. No opera- Patients with a hiatal hernia and a fibrous stric- tion can restore a damaged sphincter and surgical ture present a difficult problem. Twenty-five techniques are designed to restore and maintain (12-3%) of the 204 patients had a tight fibrous the normal anatomy of the hiatal region. stricture secondary to reflux oesophagitis (Table Recently there has been a good deal of contro- XI). They all complained of severe dysphagia; versy about the importance of the internal oeso- 10 found it difficult even to drink. In 10 the phageal sphincter in the prevention of reflux. stricture was at mid-oesophageal level, in three Some authors (Lind, Warrian, and Wankling, between 25 and 28 cm from the lower teeth, and 1966; Haddad, 1970; Cohen and Harris, 1971) in seven between 32 and 35 cm. They all needed believe that the presence or absence of oesophag- preoperative dilatations: in 16 one dilatation was itis depends only on the effectiveness of the enough, while the remaining nine required two sphincter irrespective of its position. They have or more dilatations. provided some evidence to prove that the presence or absence of hiatal hernia is of no significance TABLE XI and infer that surgery is of no value in this con- OESOPHAGEAL STRICTURE (25 PATIENTS) dition. Others (Clagett, 1966; Skinner and Belsey, Completely Partially Satisfied Satisfied Dissatisfied 1967; Collis, 1970; Ellis, 1971) believe that the 22 3 0 internal oesophageal sphincter functions best when all anatomical factors in the region of the gastro- Symptoms: Mild dysphagia 6 1 0 Barium meal: oesophageal junction are normal. The importance http://thorax.bmj.com/ Constriction ofthe oesophagus and delay of of restoring a normal anatomy at the hiatal region barium 3 1 0 has been confirmed by numerous surgeons report- Free reflux 1 1 0 ing a high operative success rate in symptomatic hiatal herniae over the past two decades. There is also experimental evidence to confirm this view. In a patient with a stricture a good result can Meiss (1963) showed that less intragastric pres- be achieved by dilatation and replacement of the sure is required to overcome the sphincter if it intra-abdominal oesophagus below the diaphragm. is located above the diaphragm than in the abdo- This may not always be easy as oesophageal men. Baue and Hoffer (1967) also found in experi- on September 23, 2021 by guest. Protected copyright. shortening is not uncommonly present. Of 25 ments on dogs that the internal oesophageal patients receiving this treatment, 22 were com- sphincter works less efficiently if placed above the pletely satisfied (although six of them still had diaphragm. residual and three were slight dysphagia) partially On evidence a definite was unsuccessful in one who present surgery has satisfied. It patient limited role in the management of hiatal had a colonic at another and although interposition hospital hernia. Radiological demonstration of a hiatal is not included. hernia (except for large para-oesophageal herniae) Results with dilatation and repair are better if is not an indication for operation. Raphael, Ellis, the stricture is in the lower oesophagus. The stric- Carlson, and Andersen (1965) reported that of ture was situated at mid-oesophageal level in six 6,571 hiatal herniae only 4% had surgical treat- of the seven patients who had residual dysphagia. ment. In Palmer's (1968) series, operation was Four of the 25 patients needed postoperative carried out on 3 8% of 1,011 patients. In our dilatations. Dilatation in fibrous strictures need series we have accepted for operation only those not be a hazardous procedure (Skinner and Belsey, patients who had such severe symptoms that they 1967). The 25 patients in this series had 59 dilata- were not prepared to continue with conservative tions without any complications. One patient with measures. Certain complications of reflux oeso- a stricture and a hiatal hernia was perforated phagitis are indications for surgery. These are during dilatation. The perforation was repaired fibrous strictures, bleeding giving rise to severe but she refused further surgery and is not included. anaemia, and obstructed or strangulated herniae. 2E Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

384 E. Hofjman and M. C. Suimner The first operation for reflux oesophagitis REFERENCES designed to restore normal anatomical and physio- Allison, P. R. (1951). Reflux oesophagitis, sliding hiatal logical conditions was described by Allison in hernia, and the anatomy of repair. Surg. Gynec. Obstet., 1951. The popularity of this procedure has 92, 419. Baue, A. E., and Hoffer, R. E. (1967). The effects of experi- declined in recent years because of reported high mental hiatal hernia and histamine stimulation on the recurrence rates (Raphael et al., 1965; Pearson intrinsic esophageal sphincter. Surg. Gynec. Obstet., 125, and Gray, 1967; Hill, 1967; Urschel and Paulson, 791. 1967). In the authors' view, Allison's concept of Brain, R. (1966). Peptic strictures of the oesophagus asso- repair remains valid and his procedure has been ciated with duodenal ulcer and operations for its relief. used in this series with slight modifications. Proc. roy. Soc. Med., 59, 929. Casten, D. F., Bernhang, A., Nach, R. J., and Spinzia, J. The two currently fashionable operative tech- (1963). A physiological basis for the surgical treatment niques are Nissen's (1961) fundoplication, where of sliding esophageal hiatus hernia. Surg. Gynec. Obstet., the fundus of the stomach is wrapped around the 117, 87. IV Clagett, 0. T. (1966). Present concepts regarding the surgical oesophagus, and Belsey's Mark repair (Skinner treatment of oesophageal hiatal hernia. Ann. roy. Coll. and Belsey, 1967) where the oesophagus is invagi- Surg. Engl., 38, 195. nated into the stomach. Nissen and Rossetti (1965) Cohen, S., and Harris, L. D. (1971). Does hiatus hernia affect found that 10% of their patients had difficulty in competence of the gastroesophageal sphincter? Newt! eructation and suffered from gastric distension. Engl. J. Med., 284, 1053. In both operations stitches are inserted into an Collis, J. L. (1970). An appraisal of the methods for treating hiatus hernia and its complications. Ann. roy. Coll. often friable oesophageal wall which may lead to Surg. Engl., 46, 338. serious complications. Nissen and Rossetti (1965) Effler, D. B. (1965). Allison's repair of hiatal hernia: late reported four deaths due to , while complication of diaphragmatic counterincision and Skinner and Belsey (1967) observed five patients technique to avoid it. J. thorac. cardiovasc. Surg., 49, 669. with oesophageal necrosis (two of whom died) Ellis, F. H. Jr. (1971). Gastroesophageal reflux. Indications following the Mark IV repair. Orringer, Skinner, for fundoplication. Surg. Clin. N. Amer., 51, 575. Haddad, J. K. (1970). Relation of gastroesophageal reflux to and Belsey (1972), reviewing hiatal hernia repairs yield sphincter pressures. , 58, 175. carried out by the Mark IV procedure, reported http://thorax.bmj.com/ Hiebert, C. A., and Belsey, R. (1961). Incompetence of the a recurrence rate of 12% in patients followed up gastric cardia without radiologic evidence of hiatal for from three to 10 years. hernia. J. thorac. cardiovasc. Surg., 42, 352. It is difficult to make valid comparisons of the Hill, L. D. (1967). An effective operation for hiatal hernia: efficacy of different surgical procedures. No single an eight year appraisal. Ann. Surg., 166, 681. method of assessment-radiological, relief of Hoffman, E. (1968). Strangulated . symptoms or patient's opinion of the operation- Thorax, 23, 541. is a reliable guide. Radiological demonstration of Lind, J. F., Warrian, W. G., and Wankling, W. J. (1966). reflux or a hernia in an asymptomatic patient is Responses of the gastroesophageal junctional zone to increases in abdominal pressure. Canad. J. Surg., 9, 32. on September 23, 2021 by guest. Protected copyright. of no great clinical significance. To the patient, Meiss, J. H. (1963). The mechanism of the cardia, an experi- all that matters is that he be cured or relieved mental study. Arch. chir. neeri., 15, 19. of his symptoms. The interpretation of sympto- Moffat, R. C., and Berkas, E. M. (1965). Bile . matic relief is personal and varies with time, and Arch. Surg., 91, 963. the patient's overall opinion of the operative result Nissen, R. (1961). Gastropexy and fundoplication in surgical may be influenced by other coincidental diseases. treatment of hiatal hernia. Amer. J. dig. Dis., 6, 954. Five of our 12 dissatisfied patients had no oeso- - and Rossetti, M. (1965). Surgery of hiatal and other phageal symptoms but suffered from a variety of diaphragmatic . J. int. Coll. Surg., 46, 663. other diseases. Olsen, A. M., and Schlegel, J. F. (1965). Motility disturbances Of 204 and caused by esophagitis. J. thorac. Surg., 50, 607. our patients with reflux oesophagitis Orringer, M. B., Skinner, D. B., and Belsey, R. H. R. (1972). hiatal herniae repaired by a modified Allison's Long-term results of the Mark IV operation for hiatal procedure, only five (2 5%) were surgical failures. hernia and analyses of recurrences and their treatment. These were all dissatisfied, four because their J. thorac. cardiovasc. Surg., 63, 25. symptoms had not improved and one because she Palmer, E. D. (1968). The hiatus hernia-esophagitis-esopha- developed dysphagia with oesophageal spasm. geal stricture complex. Amer. J. Med., 44, 566. Three of these patients also showed reflux or an Pearson, J. B., and Gray, J. G. (1967). Oesophageal hiatus hernia: long-term results of the conventional thoracic irreducible hiatal hernia. A failure rate of 2-5% operation. Brit. J. 530. is not unreasonable and this Surg., 54, operative procedure Raphael, H. A., Ellis, F. H., Carlson, H. C., and Andersen, can therefore be recommended in patients with H. A. (1965). Surgical repair of sliding esophageal hiatal severe symptoms of reflux oesophagitis. hernia: long-term results. Arch. Siurg., 91, 228. Thorax: first published as 10.1136/thx.28.3.379 on 1 May 1973. Downloaded from

A clinical and radiological review of 204 hiatal hernia operations 385

Skinner, D. B., and Belsey, R. H. R. (1967). Surgical manage- Urschel, H. C., and Paulson, D. L. (1967). Gastroesophageal ment of esophageal reflux and hiatus hernia: long-term reflux and hiatal hernia: complications and therapy. J. results with 1,030 patients. J. thorac. cardiovasc. Surg., thorac. cardiovasc. Surg., 53, 21. 53, 33. Windsor, C. W. 0. (1964). Gastro-oesophageal reflux after partial gastrectomy. Brit. med. J., 2, 1233. Toye, D. K. M., and Williams, J. A. (1965). Post-gastrectomy - and Collis, J. L. (1967). Anaemia and hiatus hernia: bile vomiting. Lancer, 2, 524. experience in 450 patients. Thorax, 22, 73. http://thorax.bmj.com/ on September 23, 2021 by guest. Protected copyright.

F